json_csv 0.0.5 → 0.0.6

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: f70535d234603ac01375388383a3823da0f73673
-  data.tar.gz: cdae7c0ddfa419662a18a097a7373e8d196a1f48
+  metadata.gz: cca6d71ea7947967ea15bfa7fd89195549e1f5b4
+  data.tar.gz: dca198bcf9db549b57a9c09bac5661194658b12e
 SHA512:
-  metadata.gz: 8793368170c8003b8bea3af8b6e84e410973589458c8ac0d244702bed2d779ee914c24c028929053ef10e4cc4ce3d8cc054324700a453d3858a95401df4d402c
-  data.tar.gz: 871e4bfcfb63b869b60f9bc413c32823f118f2bfaec4296032ea7b7bffa07876ff0db5e66cf782b2c9d76800a0fd81e8b38fd046120ee782f18390657c20fd69
+  metadata.gz: 8929b7911dbb7da2e29a56c8e6a2b93df9ab203d3202d435b3932c782654f7392b3dce1d27215021b53544864a8569a6e5b230a0bd029a21745dc27ace2e2989
+  data.tar.gz: c00fed1d0fbed0b41c0261005909290ea48990cf7b4532bc0ff60f16cbaa1251c3aaa5244f6cad0cf695d66e43f822abf9f92a881632da1acdf49143b8e37d8a

data/lib/json_csv/array_notation.rb

@@ -0,0 +1,23 @@
+module JsonCsv
+  module ArrayNotation
+    BRACKETS = 'BRACKETS'.freeze
+    DASH = 'DASH'.freeze
+
+    VALID_ARRAY_NOTATIONS = [BRACKETS, DASH].freeze
+
+    def self.bracket_header_to_dash_header(bracket_header)
+      # e.g. replace occurrences of '[1]' with '-1'
+      bracket_header.gsub(/(\[(\d+)\])/, '-\2')
+    end
+
+    def self.dash_header_to_bracket_header(dash_header)
+      # e.g. replace occurrences of '-1' with '[1]'
+      dash_header.gsub(/(-(\d+))/, '[\2]')
+    end
+
+    def self.raise_error_if_invalid_array_notation_value!(error_class, array_notation)
+      raise error_class, "Invalid array notation. Must be one of #{VALID_ARRAY_NOTATIONS.join(' or ')}." unless VALID_ARRAY_NOTATIONS.include?(array_notation)
+    end
+
+  end
+end

data/lib/json_csv/csv_to_json.rb

@@ -1,3 +1,5 @@
+require 'json_csv/array_notation'
+require 'json_csv/utils'
 require 'csv'
 
 module JsonCsv
@@ -13,19 +15,33 @@ module JsonCsv
     # presenting that row as un-flattened json.
     # This method works for large CSVs and uses very little memory
     # because it only keeps one row in memory at a time.
-    def csv_file_to_hierarchical_json_hash(path_to_csv, field_casting_rules = {})
+    def csv_file_to_hierarchical_json_hash(path_to_csv, field_casting_rules = {}, array_notation = JsonCsv::ArrayNotation::BRACKETS, strip_value_whitespace = true)
       i = 0
-      CSV.foreach(path_to_csv, headers: true) do |row_data_hash|
-        hierarchical_hash = {}
-        row_data_hash.each do |key, value|
-          next if value.nil? || value == '' # ignore nil or empty string values
-          put_value_at_json_path(hierarchical_hash, key, value, field_casting_rules)
+      CSV.foreach(path_to_csv, headers: true, header_converters: lambda { |header|
+        if array_notation == JsonCsv::ArrayNotation::DASH
+          JsonCsv::ArrayNotation.dash_header_to_bracket_header(header).strip
+        else
+          header.strip
         end
-        yield hierarchical_hash, i
+      }) do |row_data_hash|
+        yield csv_row_hash_to_hierarchical_json_hash(row_data_hash, field_casting_rules, strip_value_whitespace), i
         i += 1
       end
     end
 
+    def csv_row_hash_to_hierarchical_json_hash(row_data_hash, field_casting_rules, strip_value_whitespace = true)
+      hierarchical_hash = {}
+      row_data_hash.each do |key, value|
+        next if value.nil? || value == '' # ignore nil or empty string values
+        put_value_at_json_path(hierarchical_hash, key, value, field_casting_rules)
+      end
+      # Clean up empty array elements, which may have come about from CSV data
+      # that was 1-indexed instead of 0-indexed.
+      JsonCsv::Utils.recursively_remove_blank_fields!(hierarchical_hash)
+      JsonCsv::Utils.recursively_strip_value_whitespace!(hierarchical_hash) if strip_value_whitespace
+      hierarchical_hash
+    end
+
     # For the given obj, puts the given value at the given json_path,
     # creating nested elements as needed. This method calls itself
     # recursively when placing a value at a nested path, and during

data/lib/json_csv/json_to_csv.rb

@@ -7,11 +7,16 @@ module JsonCsv
     # strings (because CSVs are dumb and don't store info about data types)
     # Set first_index to 1 if you want the first element in an array to
     #
-    def json_hash_to_flat_csv_row_hash(json_hash)
+    def json_hash_to_flat_csv_row_hash(json_hash, array_notation = JsonCsv::ArrayNotation::BRACKETS)
       flat = flatten_hash(json_hash)
       # Convert values to strings because in the CSV file, all values are strings
       flat.each { |key, val| flat[key] = val.nil? ? '' : val.to_s }
-      flat
+      # If we're using dash array notation, convert the headers
+      if array_notation == JsonCsv::ArrayNotation::DASH
+        Hash[flat.map { |key, val| [JsonCsv::ArrayNotation.bracket_header_to_dash_header(key), val] }]
+      else
+        flat
+      end
     end
 
     # This method calls itself recursively while flattening a hash, and during
@@ -20,7 +25,7 @@ module JsonCsv
       if obj.is_a?(Hash)
         obj.each do |key, val|
           if key_contains_unallowed_characters?(key)
-            raise ArgumentError, 'Cannot deal with hash keys that contain "[" or "]" because these are used for internal processing.'
+            raise ArgumentError, 'Cannot deal with hash keys that contain "[" or "]" or "." because these characters have special meanings in CSV headers.'
           end
           path = parent_path + (parent_path.empty? ? '' : '.') + key
           flatten_hash(val, path, flat_hash_to_build)

data/lib/json_csv/utils.rb

@@ -0,0 +1,64 @@
+module JsonCsv
+  module Utils
+
+    # Returns true for empty strings, empty arrays, or empty hashes.
+    # Also returns true for strings that only contain whitespace.
+    # Returns false for all other values, including booleans and numbers.
+    def self.removable_value?(value)
+      return true if value.respond_to?(:empty?) && value.empty? # empty string, empty array, or empty hash
+      return true if value.is_a?(String) && value.strip.empty? # string that only contains whitespace
+      return true if value.nil?
+      false
+    end
+
+    def self.recursively_remove_blank_fields!(hash_or_array)
+      return if hash_or_array.frozen? # We can't modify a frozen value, so we won't.
+
+      if hash_or_array.is_a?(Array)
+        # Recurse through non-empty elements
+        hash_or_array.each do |element|
+          recursively_remove_blank_fields!(element) if element.is_a?(Hash) || element.is_a?(Array)
+        end
+
+        # Delete blank array element values on this array level (including empty object ({}) values)
+        hash_or_array.delete_if do |element|
+          removable_value?(element)
+        end
+      elsif hash_or_array.is_a?(Hash)
+        hash_or_array.each_value do |value|
+          recursively_remove_blank_fields!(value) if value.is_a?(Hash) || value.is_a?(Array)
+        end
+
+        # Delete blank hash values on this hash level (including empty object ({}) values)
+        hash_or_array.delete_if do |_key, value|
+          removable_value?(value)
+        end
+      else
+        raise ArgumentError, 'Must supply a hash or array.'
+      end
+
+      hash_or_array
+    end
+
+    # Recursively goes through an object and strips whitespace,
+    # modifying the object's nested child hashes or array.
+    # Note: This method modifies hash values, but does not
+    # modify hash keys.
+    def self.recursively_strip_value_whitespace!(obj)
+      if obj.is_a?(Array)
+        obj.each do |element|
+          recursively_strip_value_whitespace!(element)
+        end
+      elsif obj.is_a?(Hash)
+        obj.each_value do |value|
+          recursively_strip_value_whitespace!(value)
+        end
+      elsif obj.is_a?(String)
+        obj.strip!
+      end
+
+      obj
+    end
+
+  end
+end

data/lib/json_csv/version.rb

@@ -1,6 +1,6 @@
 module JsonCsv
 
-  VERSION = '0.0.5'.freeze
+  VERSION = '0.0.6'.freeze
 
   def self.version
     VERSION

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: json_csv
 version: !ruby/object:Gem::Version
-  version: 0.0.5
+  version: 0.0.6
 platform: ruby
 authors:
 - Eric O'Hanlon
@@ -88,8 +88,10 @@ extra_rdoc_files: []
 files:
 - README.md
 - lib/json_csv.rb
+- lib/json_csv/array_notation.rb
 - lib/json_csv/csv_to_json.rb
 - lib/json_csv/json_to_csv.rb
+- lib/json_csv/utils.rb
 - lib/json_csv/version.rb
 - lib/tasks/json_csv.rake
 - lib/tasks/json_csv/ci.rake